Modern databases are capable of holding significant amounts of data which is often structured into several tables. In order to maintain satisfactory manageability, performance and availability of data in the database, tables having a significant amount of data are generally divided into several partitions with each partition holding data falling within a predefined unique data range. Failure to regularly partition database tables can be problematic. For example, a back office database (BODB) of a client-server based gaming system typically receives several terabytes of data each week and, accordingly, if regular partitioning of database tables does not occur, the performance of the database will quickly deteriorate.
However, partitioning of databases is currently carried out manually by database administrators, typically once per hour for a client-server based gaming system. This imposes a significant workload on the database administrators.